에이전트 및 멀티 에이전트

LlamaIndex

run-llama/llama_index

LlamaIndex is an open-source Python data framework for building LLM applications with private data, offering data connectors, indexing structures, and advanced retrieval interfaces. It is complemented by LlamaParse, an enterprise platform for agentic OCR, parsing, extraction, and indexing.

공식 저장소 열기 ↗ 프로젝트 홈페이지 ↗
★ 51.4K별점
⑂ 7.9K포크 수
599미해결 이슈
Python언어
MIT라이선스
Q@project.QualityScore편집 점수

프로젝트 스크린샷

LlamaIndex 스크린샷 LlamaIndex 스크린샷 LlamaIndex 스크린샷 LlamaIndex 스크린샷 LlamaIndex 스크린샷 LlamaIndex 스크린샷

개요

LlamaIndex is designed to augment large language models with custom private data. It provides tools to ingest data from various sources (APIs, PDFs, SQL, etc.), structure it into indices or graphs for LLM consumption, and expose an advanced retrieval and query interface. The framework supports both high-level APIs for rapid prototyping and lower-level APIs for deep customization of data connectors, indices, retrievers, and query engines. It is structured around a core package with over 300 optional integration packages available via LlamaHub.

주요 기능

  • Data connectors for ingesting APIs, PDFs, docs, SQL, and more
  • Data structuring via indices and graphs for LLM use
  • Advanced retrieval and query interface
  • Over 300 integration packages on LlamaHub
  • High-level and low-level APIs for varied user expertise
  • LlamaParse integration for agentic OCR and document parsing
  • Support for multi-agent workflows
  • In-memory and disk-persisted storage contexts

요구 사항, 설치 및 빠른 시작

There are two primary installation methods. Starter: install the `llama-index` package which includes core and a selection of integrations. Customized: install `llama-index-core` and then add specific integration packages as needed (e.g., `pip install llama-index-core`, `pip install llama-index-llms-openai`, `pip install llama-index-llms-ollama`, `pip install llama-index-embeddings-huggingface`).

사용 정보

To build a simple vector store index, load documents using `SimpleDirectoryReader`, create an index with `VectorStoreIndex.from_documents(documents)`, and query it using `index.as_query_engine()`. Non-OpenAI LLMs can be configured via the `Settings` object, setting custom LLM, tokenizer, and embedding model instances before creating the index.

모델 호환성 및 사용 사례

Supports OpenAI models out of the box. Compatible with non-OpenAI LLMs such as models hosted via Ollama (e.g., `llama-3.1:latest`) and HuggingFace embedding models (e.g., `BAAI/bge-small-en-v1.5`).

라이선스 및 위험 참고 사항

The repository is licensed under the MIT license.

Editorial verification 2026-08-09: repository URL, owner, description, license and repository statistics were reviewed. License metadata: MIT. README was fetched for the channel draft; re-check repository dependencies, releases and model terms before production use.

릴리스 및 유지 관리

Not stated in the repository metadata.

Firecrawl

firecrawl/firecrawl

★ 161.1KTypeScript

LangChain

langchain-ai/langchain

★ 143.6KPython

RAGFlow

infiniflow/ragflow

★ 86.7KGo